The ethics of not voting

Eric Crampton and I see politics from roughly the same GMU-trained, public-choice-heavy, anarcho-curious perspective, so it is no surprise that we are both practitioners of conscientious non-voting. Nevertheless, we disagree about the details; here is Eric’s case for conscientious non-voting. The purpose of this post is to express my ethical argument against voting and to persuade Eric that his claim that voting constitutes contractarian consent is both erroneous and unnecessary to establish non-voting as the best ethical alternative.

Ironically, the basis for conscientious non-voting is well expressed in Jason Brennan’s essay The Ethics of Voting (and presumably in his book of the same title). Brennan argues (rightly) that there is no duty to vote, but (more controversially) that if you are going to vote, you have a duty to vote well, in a manner that you justifiedly believe will promote the common good.

What is controversial about that second claim? It starts with this: in voting, there is not the usual link between action and outcome. As Brennan writes,

In a large-scale election, such as a congressional election in the United States, the probability that an individual vote will decide the outcome of the election is vanishingly small. You are much more likely to win Powerball multiple times in a row than to cast a vote that changes the outcome of a presidential or congressional election.

Consequently, on a purely instrumental basis, it does not matter in the least how you vote. Voting accomplishes nothing, so it is not immediately obvious how voting well or voting badly could have much ethical force. Indeed, if you are an act consequentialist, this is the end of the line: go forth and vote (or not) well or badly, for it makes no difference.

Nevertheless, most of us are not act consequentialists. For us, Brennan articulates a principle that may still have some application to our impotent voting behavior. He calls it the Clean Hands Principle:

One should not participate in collectively harmful activities when the cost of refraining from such activities is low.

For Brennan, voting badly is a collectively harmful activity. Therefore, he concludes that if you vote, you should not vote badly; you have a duty to vote well.

Here I will part ways with Brennan. Voting at all, at least in a political context, is a collectively harmful activity (in other contexts, e.g., you and four friends vote to decide where to eat dinner, it is not collectively harmful). In what way is it collectively harmful? Maybe it’s best to quote Brennan again:

Bad choices at the polls can destroy economic opportunities, produce crises that lower everyone’s standards of living, lead to unjust and unnecessary wars (and thus to millions of deaths), lead to sexist, racist, and homophobic legislation, help reinforce poverty, produce overly punitive criminal legislation, and worse.

How do we know that bad choices at the polls can do those things? We have evidence from every single election. Those policies that destroy economic opportunities, that produce crises, the unjust and unnecessary wars themselves, and so on—they are popular. They are the ideas that win every time.

What Brennan misses (at least in the essay; I have not read the book) is that the cause of all these harms is not just that voters make bad choices. That is a narrow perspective. It is that so many domains are subject to collective choice in the first place. The correct response to the question, “Shall we pass a law that destroys the lives of people who use drugs, especially if they are black?” is not merely “No,” but “Take your democracy and shove it.” Merely responding “No” is collectively harmful, because it fails to challenge the implicit proposition that the domain is rightly subject to collective choice.

When you vote in an election on an issue (or for candidates who can decide an issue) that should not be subject to collective choice in the first place, your vote makes no instrumental difference. It is therefore costless not to participate. By the Clean Hands Principle, you should not vote in such an election.

I believe that Eric would more or less agree with all of the above, but if I understand him correctly, he adds a superfluous element to his argument. That element is contractarianism. If I vote on an issue that is not rightly in the collective domain, does my vote help make it rightly subject to collective choice? Does my vote, my “getting my hands dirty,” constitute consent in contractarian terms?

I think it clearly does not. The simplest example is one in which dissenters misunderstand how improbable decisive voting is. Suppose that the question I cited earlier, “Shall we pass a law that destroys the lives of people who use drugs, especially if they are black?” is in fact at issue in an election. A minority of 30% oppose this measure, and half of these correctly believe drug use should not even be subject to collective choice. Nevertheless, the 15% of people who oppose collective choice for drug use incorrectly believe that there is a significant chance that their vote will be decisive. Since the vote is happening no matter what, they show up at the polls to try to avert disaster. Does their voting on the question on the basis of mistaken beliefs about their probability of success constitute consent? I believe it does not.

Another example is the case of the Hail Mary pass; it involves no mistaken beliefs. Suppose you arrive at home to discover that some gunmen have broken into your house and are about to execute your daughter. The gunmen offer you a proposition. They happen to have 100 dice with them; if you roll 100 1s, they will spare your daughter’s life. Now, you know very well that the odds of rolling 100 1s with 100 d6 is 1 in 6¹⁰⁰, a very large number, but you say a Hail Mary and roll them anyway. Are you therefore consenting to your daughter’s execution, or at least to the proposition that rolling dice is a legitimate way of deciding whether your daughter should be executed? Again, I think not.

Like most people, I believe that consent can be a source of obligation, including political obligation. But only real consent counts; if your disgust at political discourse and need for self-expression overcomes your desire to keep your Hands Clean, then yes, your hands are dirty, but no, you haven’t consented. Consent has to be intended. The fact that contractarianism in practice relies so heavily on unintended forms of (fake) consent means that we don’t really have to take it seriously as a source of political obligation.